Guide Compliance Cybersecurity

NIS2: requirements, applicability and the 10 measures in Article 21

NIS2 sets cybersecurity risk-management and incident reporting requirements for certain entities in the European Union. Whether it applies depends on the sector, the type of service, the size of the organisation and specific inclusion rules. It does not apply in the UK, but UK organisations can be affected through EU operations, certain digital services offered in the EU or contractual requirements from EU customers. This guide explains how to check applicability, assess the ten areas of measures in Article 21 and organise the evidence you need.

1 May 2026
Updated 18 November 2026
15 min read
Key takeaways
  • NIS2 does not apply in the UK; it matters to UK organisations with EU establishments, certain digital services offered in the EU, or EU customers that pass requirements down by contract
  • Scope depends on sector, entity type and size under the EU SME definition; only specific categories are covered regardless of size
  • Article 21(2) lists ten areas of risk-management measures, and each area can require several controls, procedures and pieces of evidence
  • For a significant incident: early warning within 24 hours and notification within 72 hours of becoming aware, and a final report within one month of the notification
  • Management bodies must approve and oversee the measures and follow training; liability for infringements is set by national law

Why NIS2 is a broader obligation than NIS1

NIS2, Directive (EU) 2022/2555, replaced the original NIS Directive. Under NIS1, Member States identified operators of essential services themselves. NIS2 instead lists 18 sectors in Annexes I and II and generally covers medium-sized and large entities of the listed types, so far more organisations fall within it. Member States had to transpose it by 17 October 2024 and apply the measures from 18 October 2024. Several were late, so the rules that apply to you are those in the national law of each country where you are in scope.

NIS2 also harmonises more of the substance: minimum areas of risk-management measures (Article 21), staged reporting of significant incidents (Article 23), duties for management bodies (Article 20), and supervision and enforcement (Articles 31 to 37). Member States must provide for administrative fines with a maximum of at least €10 million or 2% of total worldwide annual turnover, whichever is higher, for essential entities, and at least €7 million or 1.4% for important entities (Article 34).

10
areas of risk-management measures listed in Article 21(2)
€10M
minimum level of the maximum fine for essential entities, or 2% of worldwide turnover if higher
24h
for the early warning after becoming aware of a significant incident

Does NIS2 apply to your organisation?

Headcount alone does not decide it. Under Article 2(1), NIS2 generally applies to public or private entities of a type listed in Annex I or II that qualify as medium-sized enterprises, or exceed the medium-sized ceilings, under the EU SME definition in Recommendation 2003/361/EC, and that provide services or carry out activities in the EU.

That definition combines staff numbers with financial thresholds. An enterprise is small if it has fewer than 50 staff and an annual turnover or balance sheet total of no more than €10 million. It is medium-sized if it has fewer than 250 staff and an annual turnover of no more than €50 million or a balance sheet total of no more than €43 million. Figures of partner and linked enterprises, such as other group companies, can count towards these thresholds. A company with 50 employees is therefore not automatically covered: it must also be of a listed type, and the thresholds are applied together.

Annex I covers sectors of high criticality: energy, transport, banking, financial market infrastructures, health, drinking water, waste water, digital infrastructure, ICT service management (business-to-business), public administration and space. Annex II covers other critical sectors: postal and courier services, waste management, chemicals, food, manufacturing (including medical devices, computers and electronics, machinery and motor vehicles), digital providers (online marketplaces, search engines and social networking platforms) and research.

Entities covered regardless of size

Article 2(2) to (4) sets out specific exceptions to the size rule. Regardless of size, NIS2 covers providers of public electronic communications networks or publicly available electronic communications services, trust service providers, top-level domain (TLD) name registries, DNS service providers and entities providing domain name registration services. It also covers entities that a Member State identifies because they are the sole provider of an essential service, because a disruption could significantly affect public safety, security or health or create a systemic risk, or because they are critical at national or regional level, as well as certain public administration entities and critical entities under Directive (EU) 2022/2557.

Cloud computing, data centre and content delivery network providers (digital infrastructure) and managed service and managed security service providers (ICT service management) are not covered automatically. They fall within NIS2 when they meet the size criteria or are identified under one of the special rules. If you provide these services, the size test and any national identification still matter.

Essential or important entity

Article 3 gives the general starting point: entities of an Annex I type that exceed the medium-sized ceilings are essential, and other Annex I or II entities within scope are important. There are exceptions. Qualified trust service providers, TLD name registries and DNS service providers are essential regardless of size, medium-sized providers of public electronic communications are essential, and Member States can identify further entities as essential or important. Classification is therefore not automatic: check it against national law and any identification by the authority. It determines the supervisory regime, with ex ante and ex post supervision for essential entities and ex post supervision for important entities.

Financial entities covered by DORA, Regulation (EU) 2022/2554, should start from DORA. Under NIS2 Article 4, where a sector-specific EU act requires cybersecurity risk-management measures or incident notification that are at least equivalent in effect, the corresponding NIS2 provisions do not apply, and Recital 28 identifies DORA as such an act. That does not mean running two duplicate programmes: assess DORA first and check NIS2 only where it still applies, for example to group entities outside the financial sector. Our DORA gap analysis guide and DORA in financial services cover that assessment.

UK organisations

NIS2 does not apply in the UK, where the Network and Information Systems Regulations 2018 remain in force. The Cyber Security and Resilience Bill would update that regime; as of October 2026 it was in the House of Lords and had not received Royal Assent. A UK company is not subject to NIS2 merely because it is outside the EU or has EU customers. It can be affected through an EU establishment that is itself in scope, or as a provider of certain digital services in the EU, such as DNS, cloud, data centre, content delivery network, managed and managed security services, online marketplaces, search engines and social networking platforms: if such a provider meets the scope criteria, Article 26 assigns jurisdiction and, without an EU establishment, it must designate a representative in the EU. Requirements that EU customers include in contracts, because they must manage supply chain security, are a separate matter: they are obligations to the customer, not direct NIS2 obligations.

An applicability-first assessment

Establish applicability before you map controls. The steps below give a workable sequence and the evidence to keep at each step.

1
Jurisdiction
Where you are established and where you provide services in the EU. For certain digital providers, the Member State of main establishment or of the EU representative (Article 26). Evidence: list of legal entities, service and customer locations, main establishment analysis.
2
Entity category
Which Annex I or II entity type, if any, each legal entity and service matches. Evidence: service catalogue mapped to the annex definitions, with the reasoning for each match or exclusion.
3
Size and classification
Staff, turnover and balance sheet under Recommendation 2003/361/EC, including partner and linked enterprises; essential or important under Article 3 and any national identification. Evidence: financial statements, group structure, classification note, any decision by the authority.
4
National obligations
The transposition law in each relevant Member State: registration, competent authority, reporting channel and any national additions. Evidence: register of applicable laws, registration confirmations.
5
Security measures
Coverage of the ten areas in Article 21(2), plus Implementing Regulation (EU) 2024/2690 where it applies. Evidence: risk analysis, approved policies, control descriptions, test results.
6
Incident notification
Significant-incident criteria, decision roles and reporting routes for each Member State. Evidence: classification procedure, decision log, exercise records.
7
Governance
Approval, oversight and training by the management body under Article 20. Evidence: approval decisions, periodic reports to the management body, training content and participants.
8
Verification
Whether the controls work and the evidence is complete, checked independently of the people who run them. Evidence: internal audit or gap assessment report, remediation plan with owners and deadlines.

Entities in the categories covered by Implementing Regulation (EU) 2024/2690, namely DNS service providers, TLD name registries, cloud computing, data centre and content delivery network providers, managed service and managed security service providers, trust service providers, and providers of online marketplaces, online search engines and social networking platforms, face additional detailed technical and methodological requirements for each measure area, and specific criteria for significant incidents.

Example: two suppliers in different positions

A UK company with 120 staff develops software that Dutch hospitals install and run on their own systems. It has no EU establishment and does not provide one of the listed digital infrastructure or ICT service management services, so it is generally not an NIS2 entity. Its hospital customers, however, must manage supply chain security and will ask for security clauses, incident notification to them and evidence of controls. If the company hosted the software as a cloud service for EU customers, it would need to check whether it is a cloud computing service provider within scope, including the Article 26 representative rule.

A Dutch managed services provider with 80 staff and €15 million turnover runs IT environments for business customers. It is an Annex I ICT service management entity above the small-enterprise thresholds, so it is directly covered by the Dutch Cyberbeveiligingswet, generally as an important entity unless identified as essential. Registration, the security measures, incident reporting, management body duties and Implementing Regulation 2024/2690 all apply to it.

The ten areas of risk-management measures in Article 21(2)

Article 21(1) requires essential and important entities to take appropriate and proportionate technical, operational and organisational measures to manage the risks to the network and information systems they use for their services, based on an all-hazards approach. Article 21(2) lists ten areas that these measures must at least include. Each area can require several controls, procedures, responsibilities and pieces of evidence, so use the list as an assessment structure rather than a ten-item checklist. What is proportionate depends on your exposure to risk, your size, and the likelihood and severity of incidents.

1
(a) Policies on risk analysis and information system security
A risk analysis covering the systems used to provide your services, with approved policies and defined review intervals.
2
(b) Incident handling
Detection, analysis, containment and recovery with defined roles, a tested response plan and criteria for significant incidents.
3
(c) Business continuity, such as backup management and disaster recovery, and crisis management
Backups and recovery plans with defined objectives, tested in practice, plus crisis management arrangements.
4
(d) Supply chain security
Security aspects of relationships with direct suppliers and service providers, taking into account their vulnerabilities, product quality and secure development practices (Article 21(3)).
5
(e) Security in acquisition, development and maintenance, including vulnerability handling and disclosure
Secure development and change practices, patch management and a documented process for handling and disclosing vulnerabilities.
6
(f) Policies and procedures to assess the effectiveness of the measures
Testing, internal audits, reviews and metrics, with results reported to the management body.
7
(g) Basic cyber hygiene practices and cybersecurity training
Baseline hygiene such as patching, secure configuration and malware protection, and training for staff. Training for members of the management body is a separate duty under Article 20.
8
(h) Policies and procedures on cryptography and, where appropriate, encryption
A cryptography policy with key management, and encryption applied where the risk analysis shows it is appropriate.
9
(i) Human resources security, access control policies and asset management
Joiner, mover and leaver processes, least-privilege access and an up-to-date asset inventory.
10
(j) Multi-factor or continuous authentication and secured communications, where appropriate
The text requires multi-factor or continuous authentication solutions, secured voice, video and text communications and secured emergency communication systems, where appropriate. Document where you apply them and why.

For area (j), good practice goes further than the statutory wording: for privileged and other higher-risk access, such as administrator accounts, remote access and access to critical systems, phishing-resistant authentication such as FIDO2 security keys or passkeys is the stronger choice. That is a recommendation, not a requirement stated in Article 21(2)(j).

Most organisations already have partial controls in these areas. Common gaps are controls that exist informally, recovery plans that have never been tested and supplier clauses that nobody monitors. For how the ten areas relate to other frameworks, see our guides to NIST CSF 2.0 and NIS2 and NIS2 versus ISO 27001.

Incident notification: deadlines for significant incidents

The reporting duties in Article 23 apply to significant incidents, not to every incident you handle internally. An incident is significant if it has caused or can cause severe operational disruption of the services or financial loss for the entity, or has affected or can affect other natural or legal persons by causing considerable material or non-material damage. National law and, for certain digital providers, Implementing Regulation (EU) 2024/2690 set more detailed criteria.

  • 24 hours: early warning. Without undue delay and within 24 hours of becoming aware of the significant incident, to the CSIRT or competent authority, indicating where applicable whether unlawful or malicious acts are suspected and whether there could be a cross-border impact.
  • 72 hours: incident notification. Within 72 hours of becoming aware: an update with an initial assessment of severity and impact and, where available, indicators of compromise.
  • On request: intermediate report. Relevant status updates when the CSIRT or competent authority asks for them.
  • One month after the notification: final report. No later than one month after the incident notification, not after awareness: a detailed description, the likely root cause or type of threat, mitigation measures and any cross-border impact. If the incident is still ongoing, a progress report is due at that point and the final report within one month of handling the incident.

An internal P1 or major incident is not automatically a significant incident under NIS2, and a lower internal priority does not rule one out. Set out the classification criteria in a documented procedure, name who decides and who reports, and record each decision with its reasons and time. Assess significance during triage rather than after the forensic investigation, and escalate to the decision-maker as soon as the criteria may be met.

Important

An early warning does not require a complete picture of the incident. It needs the information available at that point. Waiting for the investigation to finish before deciding whether to report puts the deadlines that apply to you at risk.

Management body duties under Article 20

Article 20 requires the management bodies of essential and important entities to approve the cybersecurity risk-management measures, oversee their implementation and follow training. Member States must ensure that members of management bodies can be held liable for infringements of these duties, but how that liability works is set by national law.

For essential entities, Article 32(5) allows a temporary ban on a person exercising managerial functions, but only where earlier enforcement measures have proved ineffective and a set deadline to remedy the situation has passed. An incident on its own does not prove negligence or create personal liability; what matters is whether the duties were met.

The training should give members enough knowledge and skills to identify risks and assess cybersecurity risk-management practices and their impact on the entity's services. Document the content, participants, frequency and why the training suits the entity's risks. Attendance records alone do not show this, and board minutes that note a cybersecurity discussion do not by themselves establish adequate governance. Keep evidence of approval decisions, periodic reporting, follow-up of actions and review of effectiveness.

What to do now

Start with applicability, using the assessment steps above, then assess the ten areas against the legal wording and the national law that applies to you.

  • Confirm applicability and classification for each legal entity and Member State, and complete any registration that national law requires.
  • Assess each Article 21(2) area for implementation, effectiveness and evidence, rather than ticking it off as present.
  • Prioritise gaps by operational risk, the severity of the unmet obligation, the effectiveness of existing controls and the availability of evidence. Documentation is needed to demonstrate compliance, but it does not replace implementation and testing.
  • Add a documented significant-incident classification and escalation procedure to your incident response plan, with named decision-makers and reporting routes, and test it in an exercise.
  • Review contracts with critical suppliers for security requirements, incident notification to you and rights to evidence or audit.
  • Prepare management body approval, reporting and training, with the evidence described above.

Obligations and supervision depend on national law. In the Netherlands, for example, NIS2 is implemented by the Cyberbeveiligingswet, in force since 15 August 2026. An organisation that can show a completed gap assessment and a remediation roadmap is better placed to demonstrate compliance than one that has done nothing.

The NIS2 Directive (EU) 2022/2555 and Implementing Regulation (EU) 2024/2690 are published on EUR-Lex. The NCSC publishes guidance for UK organisations on the NIS Regulations and cybersecurity practice.

Frequently asked questions

Does NIS2 apply to every company with more than 50 employees in the EU?

No. An organisation must be of a type listed in Annex I or II and generally be at least medium-sized under the EU SME definition in Recommendation 2003/361/EC, which combines staff numbers with annual turnover or balance sheet total and can include partner and linked enterprises. Some categories, such as DNS service providers, TLD name registries, trust service providers and providers of public electronic communications, are covered regardless of size, and Member States can identify others. Headcount alone does not decide applicability.

Are all cloud providers covered regardless of size?

Not automatically. Cloud computing, data centre, content delivery network, managed service and managed security service providers are covered when they are medium-sized or larger, or when a Member State identifies them under the special rules in Article 2. The categories covered regardless of size are listed in Article 2(2) to (4), for example DNS service providers and TLD name registries. Cloud providers within scope must also meet the detailed requirements of Implementing Regulation (EU) 2024/2690.

What are the NIS2 incident reporting deadlines?

For a significant incident, the entity sends an early warning within 24 hours and an incident notification within 72 hours of becoming aware of it. The final report is due no later than one month after the incident notification. The authority can ask for intermediate reports, and if the incident is still ongoing at that point, a progress report is due, with the final report within one month of handling the incident. The deadlines do not apply to every internal incident, only to those that meet the significance criteria in Article 23 and national law.

Does NIS2 apply directly to UK suppliers?

Not merely because they are outside the EU or have EU customers. A UK supplier can be directly affected through an EU establishment that is in scope, or as a provider of certain digital services in the EU, such as cloud, DNS or managed services, where Article 26 sets the jurisdiction and requires a representative in the EU. Security requirements that EU customers include in contracts are a separate obligation to those customers.

What are the ten areas of measures under NIS2 Article 21(2)?

Policies on risk analysis and information system security; incident handling; business continuity and crisis management; supply chain security; security in acquisition, development and maintenance, including vulnerability handling and disclosure; policies to assess effectiveness; basic cyber hygiene and cybersecurity training; cryptography and, where appropriate, encryption; human resources security, access control and asset management; and multi-factor or continuous authentication and secured communications, where appropriate. Each area can require several controls and pieces of evidence.

Can members of the management body be personally liable under NIS2?

Article 20 requires Member States to ensure that members of management bodies can be held liable for infringements of their duties to approve and oversee the measures, but the mechanism is set by national law. For essential entities, Article 32(5) allows a temporary ban from managerial functions only after earlier enforcement measures have failed and a deadline has passed. An incident on its own does not establish negligence or personal liability.

What is the maximum fine for NIS2 non-compliance?

Member States must provide for maximum fines of at least €10 million or 2% of total worldwide annual turnover, whichever is higher, for essential entities, and at least €7 million or 1.4% for important entities. National law can set higher maximums. Authorities can also issue binding instructions, order security audits and take other enforcement measures under national law.

Ryland Deakin
About the author
Lead Consultant, Cyvra · CISM · CompTIA Security+ · MCP

Ryland has delivered cybersecurity, compliance, and IT management programmes for regulated organisations across the UK and the Netherlands for over 20 years. View full profile

Talk to Cyvra

Not sure if you're in scope for NIS2?

We run NIS2 gap assessments for organisations in the Netherlands and UK. A clear picture of where you stand beats guessing.

Disclaimer: This article is for general informational purposes only and does not constitute legal, regulatory, or professional advice. Cyvra makes no warranty as to the accuracy or completeness of this content, which may not reflect the most current regulatory developments. Readers should seek independent legal and regulatory advice appropriate to their specific circumstances. Cyvra accepts no liability for any loss arising from reliance on this content.